Mitigating barren plateaus with transfer-learning-inspired parameter initializations

DOI10.1088/1367-2630/ACB58EMaRDI QIDQ6042412zbMATH OpenFDO

Authors Huanyu Liu, Yuchun Wu, Yongjian Han, Guo-Ping Guo

Publication date 10 May 2023

Published in New Journal of Physics (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://arxiv.org/abs/2112.10952

zbMATH Keywords

barren plateausquantum computationtransfer learningvariational quantum algorithms

Mathematics Subject Classification ID

Other nonclassical models of computation (68Q09) Quantum computation (81P68)

Abstract: Variational quantum algorithms (VQAs) are widely applied in the noisy intermediate-scale quantum era and are expected to demonstrate quantum advantage. However, training VQAs faces difficulties, one of which is the so-called barren plateaus (BP) phenomenon, where gradients of cost functions vanish exponentially with the number of qubits. In this paper, inspired by transfer learning, where knowledge of pre-solved tasks could be further used in a different but related work with training efficiency improved, we report a parameter initialization method to mitigate BP. In the method, a small-sized task is solved with a VQA. Then the ansatz and its optimum parameters are transferred to tasks with larger sizes. Numerical simulations show that this method could mitigate BP and improve training efficiency. A brief discussion on how this method can work well is also provided. This work provides a reference for mitigating BP, and therefore, VQAs could be applied to more practical problems.

Recommendations

• Plateau phenomenon in gradient descent training of RELU networks: explanation, quantification, and avoidance
• Barren plateaus from learning scramblers with local cost functions
• A theoretical framework for deep transfer learning
• Balanced joint maximum mean discrepancy for deep transfer learning
• Risk bound of transfer learning using parametric feature mapping and its application to sparse coding
• Scalable transfer learning in heterogeneous, dynamic environments

Cites work

• Hybrid quantum variational algorithm for simulating open quantum systems with near-term devices
• Random quantum circuits are approximate 2-designs
• The theory of variational hybrid quantum-classical algorithms

Cited in

(5)

• Barren plateaus from learning scramblers with local cost functions
• Impact of barren plateaus countermeasures on the quantum neural network capacity to learn
• The barren plateaus of quantum neural networks: review, taxonomy and trends
• Variational quantum eigensolver toward non-bonded interaction system with hardware-efficient ansatz
• Investigating and mitigating barren plateaus in variational quantum circuits: a survey

This page was built for publication: Mitigating barren plateaus with transfer-learning-inspired parameter initializations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6042412)